Cristiano Soares

Broadband matched-field processing: Coherent and incoherent approaches

Matched-field based methods always involve the comparison of the output of a physical model and the actual data. The method of comparison and the nature of the data varies according to the problem at hand, but the result becomes always largely conditioned by the accurateness of the physical model and the amount of data available. The usage of broadband methods has become a widely used approach to increase the amount of data and to stabilize the estimation process. Due to the difficulties to accurately predict the phase of the acoustic field the problem whether the information should be coherently or incoherently combined across frequency has been an open debate in the last years. This paper provides a data consistent model for the observed signal, formed by a deterministic channel structure multiplied by a perturbation random factor plus noise. The cross-frequency channel structure and the decorrelation of the perturbation random factor are shown to be the main causes of processor performance degradation. Different Bartlett processors, such as the incoherent processor [Baggeroer et al., J. Acoust. Soc. Am. 80, 571-587 (1988)], the coherent normalized processor [Z.-H. Michalopoulou, IEEE J. Ocean Eng. 21, 384-392 (1996)] and the matched-phase processor [Orris et al., J. Acoust. Soc. Am. 107, 2563-2375 (2000)], are reviewed and compared to the proposed cross-frequency incoherent processor. It is analytically shown that the proposed processor has the same performance as the matched-phase processor at the maximum of the ambiguity surface, without the need for estimating the phase terms and thus having an extremely low computational cost.

An experimental demonstration of blind ocean acoustic tomographya)

Despite the advantages clearly demonstrated by ocean acoustic tomography (OAT) when compared to other ocean monitoring techniques, it suffers from several technical-related drawbacks. One is the requirement for rather expensive equipment to be maintained and operated at several locations in order to obtain sufficient source–receiver propagation paths to cover a given ocean volume. This paper presents the preliminary feasibility tests of a concept that uses ships of opportunity as sound sources for OAT. The approach adopted in this paper views the tomographic problem as a global inversion that includes determining both the emitted signal and the environmental parameters, which is a similar problem to that seen in blind channel identification and was therefore termed blind ocean acoustic tomography (BOAT). BOAT was tested on a data set acquired in October 2000 in a shallow-water area off the west coast of Portugal, including both active and passive (ship noise) data. Successful results show that BOAT is abl...

Environmental inversion using high-resolution matched-field processing

This paper considers the inversion of experimental field data collected with light receiving systems designed to meet operational requirements. Such operational requirements include system deployment in free drifting configurations and a limited number of acoustic receivers. A well-known consequence of a reduced spatial coverage is a poor sampling of the vertical structure of the acoustic field, leading to a severe ill-conditioning of the inverse problem and data to model cost function with a massive sidelobe structure having many local extrema. This causes difficulties to meta-heuristic global search methods, such as genetic algorithms, to converge to the true model parameters. In order to cope with this difficulty, broadband high-resolution processors are proposed for their ability to significantly attenuate sidelobes, as a contribution for improving convergence. A comparative study on simulated data shows that high-resolution methods did not outperform the conventional Bartlett processor for pinpointing the true environmental parameter when using exhaustive search. However, when a meta-heuristic technique is applied for exploring a large multidimensional search space, high-resolution methods clearly improved convergence, therefore reducing the inherent uncertainty on the final estimate. These findings are supported by the results obtained on experimental field data obtained during the Maritime Rapid Environmental Assessment 2003 sea trial.

Adaptation of Ranaviruses from Peneda-Gerês National Park (Portugal) to Cell Cultures and their Characterization

Ranaviruses are known to produce iseases in fish, amphibians and reptiles, being involved in high mortality episodes in aquaculture farms and in the decline of amphibian populations worldwide. Mass mortality episodes were previously detected in Triturus marmoratus from two lagoons of the Peneda-Geres National Park (PGNP) (NW Portugal). Ranavirus-like particles were identified in tissues of diseased newts T. marmoratus and T. boscai . Here we report on the molecular and ultrastructural characterization of these viruses.

Matched field processing: environmental focusing and source tracking with application to the North Elba data set

Experimental results on the localization of a moving sound source in shallow water are presented. Genetic algorithms (GA) are used to first intensively estimate the environment from a fixed source part of the data. Then, assuming the stationarity of the environment, localization is carried out in the moving source part of the data. Comparison with results obtained previously on the same data set shows that the source range error is reduced by 75%, while the source depth error is reduced by approximately 50%.

Experimental testing of the blind ocean acoustic tomography concept

Acoustic focalization is a well known concept that aims at estimating source location through the adjustment of multiple environmental parameters. This paper uses the same concept for inverting water column sound speed in a blind fashion, where both source location and source emitted waveform are not known at the receiver — that is Blind Ocean Acoustic Tomography (BOAT). The results obtained with BOAT, using ship noise data received on a vertical line array in a shallow water area off the coast of Portugal, show that it is indeed possible to obtain reliable joint estimates of source location and water column sound speed. During that process, it was shown that source range and depth, and Bartlett power, where good indicators of the degree of focus of the model being used.

BLIND OCEAN ACOUSTIC TOMOGRAPHY WITH SOURCE SPECTRUM ESTIMATION

Classic acoustic tomography uses controlled sound sources to probe the ocean for its physical properties. Instead, passive ocean acoustic tomography aims at using natural noise sources, such as wind induced noise, wave noise, or shipping noise with the scope of inverting for the ocean and/or bottom geophysical properties. Most studies found in the literature make use of ambient noise and sea surface wind generated noise, to invert bottom parameters in shallow water regions. Recently, another approach used shipping noise as illuminating signals to invert for water column parameters [Jesus et al., Conference on Acoustic Variability, Lerici (Italy), September 2002]. In that work, a focalization process was used to simultaneous invert known geometrical and unknown environmental parameters. In particular it was shown that known geometrical parameters such as source range and depth, and receiving array geometry, could be used as focus and out of focus indicators. During the focus periods, estimated water column parameters favorably compared to independently measured values. One of the diculties found with the shipping noise was the low received power and the dicult y to determine a sucien t number of stable frequencies. In the present work, the received signal is used to deconvolve the source power, and thus obtain a full-spectrum weighting function for optimum frequency combination during the focalization process. Results obtained in the same ship noise data set have shown an certain improvement where a stable localization and inversion could be seen throughout the run.

Shipping noise field calibration via source inversion

There is significant evidence that the low-frequency mean acoustic noise pressure level in the ocean has been constantly increasing in the last decades. The main noise sources responsible for this increase were identified as ship traffic, offshore construction and oil & gas surveying. The Portuguese Navy funded SUBECO project aims at deploying a network of multiparametric offshore buoys for environmental and acoustic monitoring. A preferred location for the individual buoys is within or close by the ship traffic separation lanes along the west coast of Portugal. It is therefore expected that the received acoustic field will be dominated by shipping noise as a mixture of both short and long range ships. Most ships will have an AIS so their position may be known at all times. One of the objectives of this network is to use the recorded sound to infer and calibrate model predicted noise in a wider area. This work proposes a relatively simple technique for estimating the radiated noise level of ship tracks within acoustic reach of a single (or a small number of closely localized) hydrophone(s). It is expected that the data constrained predicted field with the estimated source levels will provide a better fit to the actual wide area acoustic field. This work is a contribution to fulfill the requirements of Portugal to the European Union Marine Strategy Framework Directive (MSFD) aiming at a good environmental status.

On the performance of geo-acoustic estimation for a distributed sensor array

The vision underlying the Widely scalable Mobile Underwater Sonar Technology (WiMUST) project is that of developing advanced cooperative and networked control / navigation systems to enable a large number (tenths) of marine robots towing small acoustic arrays to act as a coordinated team for seismic sub-bottom imaging. The space-time coherent processing of bottom returns requires the ensemble of short acoustic arrays to be seen as a single spatially distributed sensor array. Since the vehicles are free to move along range, cross-range and depth the resulting distributed sensor array may take, at least conceptually, any spatial shape. With array shape freedom comes the question of which is the most suitable (or optimal) array geometry for sub-bottom imaging and inversion. The answer to this question hinges, among others, on the definition of performance of a seismic sub-bottom profiling system. Determining the optimal sensor array geometry is clearly a ill-posed problem, since the optimal geometry is itself bottom dependent, and there is no such environment as “one size fits all”. This work addresses several criteria for sub-bottom profiling system performance including gain, resolution and probability of detection. Two physical models will be tested: one based on acoustic wave reflection used in traditional seismic imaging, and another normally used in matched-field bottom properties estimation, that includes propagation and refraction. Simulations to support the theoretical developments and algorithms were obtained on a scenario inspired in a real environment off the coast of Peljesac (Croatia).

Seismoacoustic bottom inversion with AUV towed streamers: A multi-stage approach

The WiMUST (Widely scalable Mobile Underwater Sonar Technology) Project envisions using a team of autonomous underwater vehicles towing short acoustic arrays for seismic surveying of seabottom geoacoustic properties. One of the objectives in the project is to tackle the inversion of acoustic data collected with short towed horizontal arrays by means of a Matched-Field Inversion (MFI) technique. While there is great deal of experience in MFI and the so-called focalization applied to horizontal propagation scenarios, in near vertical propagation scenarios, with a source receiver horizontal distance limited to a few tens of meter or less, there is little understanding in terms of feasibility of the acoustic inversion of bottom properties. In particular, the simultaneous inversion of bottom properties (soundspeeds, densities, attenuations) of multiple bottom layers needs to be tackled, since the experimenter has to account for the admissible mismatch of other environmental properties such as water soundspeed and depth, and the potential solution ambiguity inherent to an optimization problem with ten or more unknown parameters. The actual simulation study, carried out with an environmental scenario and geometric set up based on the Peljesac data set, considers a shallow water acoustic propagation scenario with a short array. A sensitivity analysis in MFI provides understanding on the observability of the unknown parameters of interest. A mismatch analysis indicates that watercolumn mismatch (soundspeed and depth) may cause the MFI procedure to break down. Based on the conclusions taken from the sensitive and mismatch analysis, an iterative acoustic inversion concept with feedback of intermediate parameter estimates is developed and tested with simulated data.